Hacker News new | past | comments | ask | show | jobs | submit login
Is Tribalism a Natural Malfunction? What computers teach us about getting along (nautil.us)
57 points by dnetesn on Sept 23, 2017 | hide | past | favorite | 46 comments



Natural Malfunction. Has humanity finally grown up so much, we can talk about the bugs the process called evolution produces?

Can we actually talk about loop-deformations, and the danger that self-optimizing systems optimize try to optimize the social environment to the configuration that allows for a maximum self-reproduction performance?

Will there be a open discussion on how only a constant economic bribe via surplus prevents civilization from sliding back into cyclic behavior?

Or will it be just findings, a cultural look up on academic behavior from the resulting emotions and then dismissal of ones own findings as something of the past to overcome by praying and wishing/ strict social controls? Even though strict social controllism may also be a strategy that evolved as part of the loop?

Who watches for bugs in the bug report mechanism? Who patches our social meta-organims against the found bugs? Who sets up laws to punnish exploits?

One needs to stop having any influences of feeling towards humanitys bugs whatsoever.

Feelings do not solve problems- at best they allow to avoid operations that crash the "cultural & biological software"- but the error persists. Just avoiding tribalism means, that it will rear its face, whenever there is stress on the system and it will do damage then, because by nature of not-understanding it will be uncontrollable.

And why not use it for good? If tribalism is used to proof holes in the theorys of the neigbouring academia clans science - with the quasi-religious fervor of soccer fans, where is that not a positive, as long as a idea is not simply dismissed because of otherness?

Imagine if you where allowed to wear shawls with dots for non-repeatable experiments in the colors of a "enemy" university?


Oh you can talk about it a lot, but do you fix it "Gattaca style" (troublesome)

> Can we actually talk about loop-deformations, and the danger that self-optimizing systems optimize try to optimize the social environment to the configuration that allows for a maximum self-reproduction performance?

That's what worked for millions of years. It's been tens of thousands of years since agriculture took over and while I don't subscribe to the Paleo ideas, our bodies evolved for lifespans long enough to reproduce, and for a low ratio of (calories obtained)/(calories spend obtaining food)


How about attention to bugs like displays of sexual fitness, that make us plunder the earth, spewing carbon dioxide as we go, in order to consume and display like bowerbirds.


Are Feelings a Natural Malfunction?


Tribalism is seen all across nature. It is an important mechanism for adaptability and humanity wouldn't have made it as far as it has without it. The article equates it with its worst manifestations (war and genocide), but fails to appreciate its ecological importance. A world with only one large society is fragile and subject to large catastrophic failure.

This is just like the issue of monoculture in software security. If there are many societies (tribes) there are more chances for survival. There are also more chances for independent discoveries of better ways of living that would otherwise be lost to groupthink.

Tribalism is a diversified portfolio. It's not a bug, it's a feature. Our challenge is not to get rid of tribalism but rather how to get the best parts of it without the worst.


I agree with your conclusive point. On the others, I'm not sure we need to worry about losing tribalism any time soon so I hope you're not losing sleep over it.


I'm not. I'm just stunned by the ignorance of people who aim for that goal.


Of course, those people are all the while often reiterating it.


Anyone who has watched a nature documentary knows that tribalism is essential for an organism longterm survival. Humans are just imaginative enough to want to transcend nature and reality.

To quote Cormac McCarthy, "There's no such thing as life without bloodshed. I think the notion that the species can be improved in some way, that everyone could live in harmony, is a really dangerous idea. Those who are afflicted with this notion are the first ones to give up their souls, their freedom. Your desire that it be that way will enslave you and make your life vacuous."


In the history of mankind (one of) the usual ways of having your tribe exterminated was meeting another tribe.

Idealist ideas of the whole humanity singing kumbaya in a circule by the fireplace are naive at best.

Doesn't mean "our tribe" (which exists in a lots of levels, it could be a sports team, it could be a national identity, or something more vague) needs to be hostile to others (or shoot first)


At the same time, many tribes benefitted from exchange and alliances. Tribalism does not necessarily imply brutal conflict.


True. There is at least one apparent exception, Australia, during hunter gatherer times. But everywhere else...

Exchange and alliance also happened; often in aid of success in brutal conflicts.


I'd like to know how much culture affects cooperation/trust in society vs other factors, such as genetics. I have a hypothesis that culture is much like a protocol. When all members of society speak the same protocol around each other, there is cooperation. The opposite is like your tragedy of the commons effect. This isn't to say that multi-culturalism would be bad if this hypothesis were true, only that all members of the society may need to at least understand a common culture to have cooperation.


The author is as quick to describe it as a "natural malfunction" as he is to his findings of his toy models to large complex societies. Can his analogy be applied on a different order of organization? Cells have very limited means of communicating with each other like this model, and in multi-cellular organism, they have a tendency to cooperate with things that share their code. Could this be said to be a malfunction?


I disagree with the term malfunction as well, it's more of an heuristic for dealing with lack of perfect information about others


Wow. So if I'm understanding the article correctly, it says that tribalism occurs both naturally and in simulations. However, even though this is the case, we should despise tribalism for some reason that is frankly not exactly clear.

does the depth of cultural self-loathing know no bound?


Sex is a natural desire, but it's probably a good thing we have laws against those who force their way onto others whenever they're horny. Just because it's natural doesn't necessarily mean it's best.


You are positing a false dichotomy here. Equivalent conclusion would be that 'sex is a natural malfunction and needs to be removed'.


> does the depth of cultural self-loathing know no bound

I'm puzzled by this (possibly rhetorical?) question. Which culture is supposed to be loathing itself here?


Something that would be called "western/christian/european" civilization


One big point of “Western civilization” (religious mores about cousin marriage and inheritance; the classical liberal / republican tradition with religious toleration and representative government; science; European legal traditions with an independent judiciary; economic liberalism; down through to the Geneva Conventions; the UN and similar institutions; the Universal Declaration of Human Rights; universal education; universal suffrage; etc.) is rejecting tribalism in favor of the rule of law and more just and egalitarian social relations and decisionmaking processes.


One big point of “Western civilization” ... is rejecting tribalism

It isn't really rejecting tribalism, it's something more akin to the 'FairBot' tribe from the article. It is a way of selecting members based on adherence to a very complex set of tendencies and behaviours; in other words, a Shibboleth. It's been more successful in the recent term because it has succeeded in identifying tribe members independent of ethnic origin. It absolutely does not shy away from rejecting people who fail at performing the Shibboleth and it doesn't care if those people are the direct descendents of tribe members. Exhibit A: the number of people in prison in the West.


> It isn't really rejecting tribalism, it's something more akin to the 'FairBot' tribe from the article.

I was going to say this exact same thing --- that, as far as the simulation understands, FairBot is just a more complicated Shibboleth.

To properly perform our Shibboleth is to have loved other people, then develop and understanding of murder/genocide and be given an opportunity to express your feeling about it. If these things don't bother you, then you've failed the Shibboleth and need to be killed (or imprisoned) so that you don't damage our society.

And, while our society certainly doesn't implement this perfectly, this is the best you can do. The shibboleth has worked so well because, unless you're a sociopath, it comes out of the root of your being. In my mind, the experience of transcendent love is so powerful that we are willing to reject ANY destruction of it, even if it doesn't directly affect us (i.e. I wouldn't kill a mother because, having a mother myself, I know it would destroy a very power human bond).


> religious mores about cousin marriage and inheritance

(just because I didn't know that word) Mores: "The accepted traditional customs and usages of a particular social group."

> is rejecting tribalism in favor of the rule of law

Agreed. You just have to convince the other "tribes" to do that as well. And the more people are forced to accept other people from a more distant (not geographical distance) culture, the stronger the backlash is

> more egalitarian social relations and decisionmaking processes

How many forced marriages, FGMs and honour killings are happening now under the auspice of western secular governments/societies?


As far as I can tell many cultures throughout the world are changing extremely rapidly in response to collapse of agrarian and nomadic societies, urbanization, education and economic opportunity for women, birth control, proliferation of communication technology, and so on. There are also many problems related to foreign corporate actors with no local accountability, air and water pollution, environmental destruction, massive distribution of small arms, the international narcotics trade, etc.

I’m not sure I understand your question. Are you talking about autocratic regimes propped up by western powers for short-sighted self-serving geopolitical reasons? Which western/secular governments and which honor killings are you talking about?


> As far as I can tell many cultures throughout the world are changing extremely rapidly in response to collapse of agrarian and nomadic societies, urbanization, education and economic opportunity for women, birth control, proliferation of communication technology, and so on.

Agreed. When they have the chance

> Are you talking about autocratic regimes propped up by western powers for short-sighted self-serving geopolitical reasons?

While propped up regimes are often undemocratic, the anti-secularism (which is different from simply "religious aligned" - like most republicans, as an example) and other factors do not come from it.

Though I agree auspice is a bad word to describe it, as governments are indifferent rather than supportive

https://en.wikipedia.org/wiki/Shafia_family_murders

http://www.bbc.co.uk/news/world-us-canada-41193899

https://www.nspcc.org.uk/preventing-abuse/child-abuse-and-ne...


Your first two links seem like isolated examples of crimes for law enforcement to handle. Your third link is about migration of women have been mutilated to England. I’m not sure what you are trying to say with it, or what it has to do with secular western government.

If we want to stop female genital mutilation in Eritrea, Ethiopia, and Somalia (where the women under discussion are migrating from), it seems to me – though I am certainly not an expert on the region – that the way forward is to stop flooding them with small arms, work on improving the local economy and its connections to the outside world, work on improving the political and economic power of local women, not turn a blind eye to ethnic violence, stop uncritically supporting the awful Saudi monarchy, work to empower respected people in the region to speak out against such practices, put pressure on local governments and courts to change their laws and enforcement, and so on.

Trying to impose cultural changes on people by external force is historically not a productive method.


> Trying to impose cultural changes on people by external force is historically not a productive method.

Correct, but tolerating cultural practices incompatible with modern societies is causing local populations to feel frustrated (not to mention the imposition of guilt onto them).


From your examples, you only view this as an issue when it's foreign cultural practices. Honor killings are an extreme form of domestic violence, do you feel the same frustration and guilt over all the ~800,000 incidents in the US every year?


> Honor killings are an extreme form of domestic violence, do you feel the same frustration and guilt over all the ~800,000 incidents in the US every year?

Criticizing one practice does not mean other practices are acceptable. So yes, domestic violence is a cause of concern. But nice whataboutism.


Hardly that I would think. The anti-tribal direction of modern world history originated in the Enlightenment, which, although catalysed by contacts with Muslim, Indian and Chinese civilisations, was nothing if not profoundly 'Western'.


How is "western/christian/european" different from "western"?

EDIT: the current framework of international order has roots in Westphalian notions of nation states, which is indeed christian and european.


>So if I'm understanding the article correctly, it says that tribalism occurs both naturally and in simulations.

My take was that the article does not quite make these claims, but uses the simulations to model implications of them.

>we should despise tribalism for some reason that is frankly not exactly clear

We should move beyond tribalism because of it's inherence of prejudice.


It does seem that the article makes these claims - it describes a particular experiment where they unexpectedly discovered that tribalism emerged as the dominant (most successful) strategy in a competitive game they simulated.


What these simulations show is that tribalism is a winning strategy. Which means not following it leads to death.

>We should move beyond tribalism because of it's inherence of prejudice.

Considering an inherent property of a winning strategy evil means your code is non-optimal and is inevitably going to disappear.


It is a 'winning strategy' only for the winners. We can just as well call it a 'losing strategy'. As the players become more intelligent, they are able to cooperate better and everyone wins, meaning cooperation is even 'more winning'. That is, unless your intentions here are something like eugenics.


It's only possible for everyone to win if basic resources are infinite. Tribalism is the only pareto optimal strategy for finite resources, especially land on Earth.


Other things that occur both naturally and in simulations: revenge, jealousy, violence, territoriality, oligarchy, thumb sucking...


Tribalism tends to mean an "us vs them" attitude with emotive, knee-jerk dislike and demonising of "them", plus a disdain for constructive criticism of "us". I think history provides plenty of evidence of this and of how toxic it is.


The irony is that the people who complain loudest about tribalism (nationalism, xenophobia) in contemporary political culture display exactly those attitudes toward people they disagree with.

Anyone who thinks tribalism can be eliminated lacks self awareness or a least a mirror.


That's definitely an issue.

Regarding your second point, it's not a matter of the level of tribalism staying the same vs completely eliminating it. I think the overall levels have been reduced over the centuries, and there are people out there that are pretty non-tribalistic in outlook.


Allowing for coordinated protocol changes, the optimal length of pre-cooperative identification code should depend on the probability of error (defector). Which after a long enough time of winning (ie. close to zero defectors) would lead to jumping into cooperation immediately. Which would make defense against defectors impossible and lead to a total collapse as they exploit the hapless cooperators.

Interesting parallels.


Interesting. It seems like what they tested here was not that each agent should get as much points as possible like in a usual prisoner dilemma, but just that they should have more than the opponent (and thus get selected for reproduction).

Not sure about the explanation in the article "This domination would last until enough errors accumulated in the code handed down between generations for dominant machines to stop recognizing each other.", shouldn't be the agents that stop recognizing the code forced out? I guess it depends on implementation details, if every agent can fight several opponents before being or not reproduced.


I wonder if it also has an efficiency metric: a shorter agreeable program is better than a longer agreeable program. Then, over time, you start failing to detect bad programs after long periods of peace.


Tribalism in the fictional AMC show The Walking Dead is a survival response. Haha. Love that show.

Cancer is largely understood as a failure of cell death mechanism--a negative mutation within the organism. Whereas the flu virus is rationally seen as part of the environment. It's effect in the body can be disastrous, but it also exercises our immune system.

I think the conclusion is a grab bag of metaphors, and a poor conclusion to an interesting discussion of model and simulation.


OK - by where is the code? This looks like something that would be great to tweak the parameters and code and experiment on your own.

There was a similar work by Axelrode, it goes back to the 80s - but it also has some results that stunningly resonate with human moral strategies. And here is some Python code: https://github.com/Axelrod-Python/Axelrod




Consider applying for YC's Fall 2025 batch! Applications are open till Aug 4

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: